Re: VST and Sched Load Balance

From: Srivatsa Vaddagiri (vatsa_at_in.ibm.com)
Date: 05/05/05

  • Next message: Andries Brouwer: "Re: 2.6.12-rc3 fails to read partition table" 
    Date:	Thu, 5 May 2005 21:45:13 +0530
To: Nick Piggin <nickpiggin@yahoo.com.au>

    On Fri, May 06, 2005 at 12:52:48AM +1000, Nick Piggin wrote:
    > Well, there are a lot of ifs and buts. Some domains won't implement
    > fork balancing, others won't do newidle balancing, wake balancing,
    > wake to idle, etc etc.

    Good point. I had somehow assumed that these are true for all domains.
    My bad ..

    >
    > I think my idea of allowing max_interval to be extended to a
    > sufficiently large value if one CPU goes idle, and shutting off all
    > CPU's rebalancing completely if no tasks are running for some time
    > should cater to both hypervisor images and power saving concerns.

    Maybe we should check with virtual machine folks on this. Will
    check with UML and S390 folks tomorrow.

    > >A possible patch for B follows below:
    > >
    >
    > Yeah something like that should do it.

    Ok ..thanks. Will include this patch in the final
    load-balance-fix-for-no-hz-idle-cpus's patch! 

    -- 
Thanks and Regards,
Srivatsa Vaddagiri,
Linux Technology Center,
IBM Software Labs,
Bangalore, INDIA - 560017
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/
  • Next message: Andries Brouwer: "Re: 2.6.12-rc3 fails to read partition table"

    Relevant Pages

    • Re: Delete scheduler SD_WAKE_AFFINE and SD_WAKE_BALANCE flags
      ... even the synchronous wakeup uses the wake balancing path ... Well the normal load balancing path treats all tasks the same, ... to maximise the number of local wakeups. ...
      (Linux-Kernel)
    • Re: Suggestion on "int len" sanity
      ... > the best balancing point between what ... "grep 'int len'" really is a decent code checker, ... part is finding the 2% real problems among the output. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: sched_setaffinity() and load balancing
      ... > it seems that using sched_setaffinitycan severely disturb load ... > balancing on recent kernels. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [PATCH] Minor scheduler fix to get rid of skipping in xmms
      ... Andrew Theurer wrote: ... >>Yeah thats probably right. ... Oh OK you're thinking about balancing across the entire NUMA. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [PATCH] Load balancing problem in 2.6.2-mm1
      ... >>around can't correct the overall imbalance and so is wasteful. ... >It's the classic fairness vs throughput thing we've argued about before. ... >more reluctantly than normal balancing. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)